A ray of light in the Murray Inquiry

Advertisement
imgres

I’m going to do a half pirouette today. Rodney Maddock, adjunct professor of economics at Monash University and vice-chancellor’s fellow at Victoria University, appears at the AFR today argue that the Murray Inquiry is addressing too-big-to-fail:

David Murray’s approach to this appears to be the correct one. When a bank fails it is important that the payments and other operations of a bank continue, to minimise the shock to the economy, but that is a very different concern from making sure the bondholders do not lose any money. Of the three types of investors who lend money to banks, depositors are likely to be protected, shareholders will lose all of their investment, and the increasing global regulatory consensus is that bondholders are likely to lose some of their investment. Murray appears to be of the view that it should be made clear to bondholders that they face some risk of losing part of their investment if a bank fails.

If bank operations can continue, and taxpayers are not called on to subsidise bondholders, the whole issue of too big to fail ceases to be an issue. If there is no implied support, there is no subsidy, and there are no funding advantages to big banks from this source.

This is known as a “bail in” regime. He goes on to argue that the cost of borrowing for the banks will rise accordingly, anywhere from 10bps to 40bps, depending upon where we are in the business cycle. This outcome is credible, and in my view likely to be at the upper end of estimates. Global markets know Australian banks are 0ver-exposed to overvalued assets. The orientation towards bail-in regimes already saw ratings agencies downgrade Australian bank sub-debt last year and similar measures in Canada have seen their banks downgraded.

Advertisement

If combined with higher and rejigged capital charges that reduce and re-weight the internal risk models that feed the mortgage monster, it would go a long way towards illustrating that the inquiry and nation is taking the moral hazard in our offshore borrowing seriously. Indeed, if it were to be substantially adopted I’ll owe David Murray and his inquiry a big, fat apology, which I’ll gleefully deliver.

However, at this stage the range of proposals is so broad that it could go either way. There are also questions about how bail-ins actually function given creditors have a lot of power when the bank in question is still trying to borrow from them.

Here is more from the relevant section of the preliminary Murray document:

Advertisement

Global history records governments of all political persuasions using taxpayer funds to support distressed institutions. As undesirable as it may be to put taxpayer funds at risk to support financial institutions, in the midst of a crisis it is often the fastest and most certain option to stabilise the system and avoid widespread economic damage.

Investors can rationally surmise that the government is likely to rescue systemically important institutions if no other options exist, as their collapse would cause the most damage to the financial system and broader economy. This leads to a belief that some institutions are too-big-to-fail — that they receive an implicit government guarantee.

Perceptions of this implicit guarantee have costs. A government may need to rescue a troubled institution in a crisis, putting taxpayer funds at risk. It may also cause ‘moral hazard’. This means it may encourage systemically important institutions to take on more risk than is optimal, since they believe they receive any benefits from the risk taking while the government will bear the cost of failure. Further, investors may believe they will not make a loss, even if the institution fails, so they have less incentive to monitor the institution’s risks and apply market discipline. This can lead to a lower cost of funding for these institutions.11 Any lower funding costs might allow the institutions to become larger and more systemically important. The overall system can therefore become larger than is economically efficient, exacerbating the size of the potential cost of a crisis and therefore the size of the perceived guarantee.

Preliminary assessment

Advertisement

During the GFC, a number of countries provided financial support to struggling financial institutions, including capital injections and debt guarantees.12 The goal was to avoid wider systemic impacts if these institutions were to fail, as happened in the case of Lehman Brothers. However, in doing so these governments entrenched a belief that some institutions were too-big-to-fail. The challenge since the crisis has been to alter these beliefs.

Government can take measures to make it more likely, or more credible, to be able to impose losses on creditors, avoiding future Government interventions and lowering the cost or probability of a large institution failing. No single measure is a complete solution, but each strengthens the likelihood or credibility of orderly resolution with minimal taxpayer support, which in turn reduces the contingent liability to the Government and perceptions of an implicit guarantee.

Policy options for consultation

It is hard to completely eliminate perceptions that some institutions are too-big-to-fail, as there will always be pressure on governments to prevent the disorderly failure of a financial institution. However, the Government can take measures to minimise the extent of these perceptions through making an orderly resolution more likely with minimal need for Government support, and reducing the probability that such institutions will fail.

Globally, the GFC revealed: complexity and interconnectedness was greater than appreciated; many global financial institutions had too little capital to withstand a large shock; moral hazard was prevalent; liquidity can disappear in a crisis; and there was a lack of focus on system-wide risks. International forums, particularly the G20, have taken these lessons and sought international policy responses to reduce the potential for taxpayer funds being put at risk from government support for systemically important financial institutions (SIFIs). Among other things, this has resulted in strengthened international frameworks for capital, liquidity, financial market infrastructure and resolution.

Australia has been part of this international process, including through its 2014 Presidency of the G20, and has adopted many of the measures. However, other countries have pursued a range of additional steps — or gone further than Australia — to reduce the potential costs posed by systemically important institutions. These additional measures are worth examining in the Australian context:13

  • Further strengthening recovery and resolution frameworks
  • Further enhancing regulatory requirements for systemic financial institutions, including higher capital requirements and stronger risk management and stress testing requirements
  • Mandating structural changes for individual banks, or imposing general ring-fencing requirements or banning certain activities considered to be high risk

The options discussed aim to reduce the costs associated with too-big-to-fail institutions. They aim to make the financial system safer and reduce the likelihood of taxpayer funds being used to support a financial institution. Many of these measures could also have an effect on competition.

Some of these options would be relatively straightforward to implement with minimal cost. However, others are much more difficult to adopt and would require significant change to the Australian financial system.

Recovery and resolution preparedness

It is not possible to eliminate failure from the financial sector. It is not even desirable; the ability of good institutions to prosper and inefficient ones to fail is a key feature of competition. However, liquidating financial institutions is a complex and slow process, which can weaken confidence and lead to contagion in the broader financial system. Financial institutions need a strong, effective and credible recovery and resolution regime to help ensure any failure is orderly and has a minimal cost to the financial system, the broader economy and the Government.

Many recovery and resolution options have little or no compliance cost for industry; for example, powers are only relied upon where an institution is facing acute financial distress. In normal times, the regulatory burden of these options can therefore be negligible or non-existent. However, some options, such as requirements for industry to structurally preposition, do impose costs.

Imposing losses on creditors

When a business fails, it would ordinarily enter corporate insolvency or administration procedures to be sold or liquidated, with any value returned to creditors. Generally, the value of the assets will not be enough to repay all creditors the whole amount owing, and some will take a loss. Achieving this in a manner that meets financial stability objectives is more difficult in the context of a financial institution. Critical services provided by the institution may need to be continued or wound down in an orderly manner outside normal insolvency processes. In addition, creditors are often other financial institutions — imposing losses on these institutions, especially in the middle of a financial crisis, can worsen the situation. Disorderly resolution of one institution can create instability through a loss of confidence and changes in investor risk appetite. In many past instances, both in Australia and elsewhere, this has led governments to intervene to restore stability.

Introducing credible ways to impose losses on creditors in the event of failure assists in achieving orderly resolution with minimal use of taxpayer funds. This goes some way to addressing perceptions that some institutions have an implicit guarantee by reducing expectations of Government support, and encouraging investors to pay greater attention to risk. This is evident in Moody’s decision to place Canadian banks on a negative outlook, which was “taken in the context of previously announced plans by the Canadian government to implement a ‘bail-in’ regime”, which “may reduce [Moody’s] systemic support assumptions”.14

There are complexities involved in making it more credible to impose losses on the creditors of financial institutions. These include: questions around the nature of the liabilities that may best be able to absorb losses in resolution; ensuring that relevant creditors are capable of bearing loss without systemic contagion; appropriate mechanisms and triggers for imposing losses on creditors; interaction with the existing regulatory capital framework;15 and the effect on funding costs in normal times.

The G20 continues to consider elements of these issues, in particular through its work on the adequacy of global systemically important institutions’ loss absorbing capacity when they fail (gone concern loss absorbing capacity, or GLAC). There is value in an internationally consistent approach to promote a level playing field globally. The proposals include government and regulator discretion to impose losses on particular classes of creditors, and mandating issuance of financial instruments that resolution authorities can confidently expose to loss or convert to equity in resolution, while minimising financial instability and risks to other public interest objectives. The proposals seek to ensure resources are available to provide solvency to a systemic institution via a bridge or a bail-in transaction sufficient to sustain its critical services until the institution can be subjected to an orderly wind-down or solvent restructuring. Although these proposals are being developed for globally systemically important banks, it is possible that they will have implications for Australian institutions.

Advertisement
Advertisement

Financial market infrastructure oversight and resolution

FMI includes trading platforms, high-value payment systems, clearing and settlement systems, and trade repositories. Such facilities are critical in a modern financial system, as they match buyers and sellers and facilitate the transfer of both funds and title to assets, and thus trading of goods, services and financial assets. If these facilities fail, it can cause severe systemic disruption.

FMIs are regulated and supervised by the RBA and ASIC. Broadly, the RBA is responsible for ensuring the stability and safety of FMIs, while ASIC is responsible for their fair and effective provision of services.

The Council of Financial Regulators (CFR) has identified a number of gaps in the regulatory framework for FMIs, particularly related to resolution.16 These include: a lack of appropriate direction powers; the ability to step in to control an FMI, if appropriate; the ability to mandate the location of services; and lack of fit and proper standards for FMI directors and officers. A large focus of these proposed changes is ensuring that, if an FMI fails, regulators are able to step in to keep its critical services operating.

The Inquiry understands that CFR agencies have developed a set of legislative proposals for future Government consideration to address the gaps identified, and supports this process.

Resolution powers

International Monetary Fund (IMF) and FSB analysis concluded that Australia’s resolution regime for banks and insurers was broadly consistent with international best practice.17 However, a comparison with the FSB’s Key Attributes of Effective Resolution revealed several gaps in resolution tools and powers.

The gaps identified include: powers to address a distressed foreign bank branch in Australia; the ability to require restructuring of a regulated entity to facilitate resolution; deficiencies in powers to resolve group distress; a lack of statutory ‘bail-in’ powers to impose losses on particular creditors; no resolution or privately funded protection funds; and no formal mechanism for recovery of taxpayer funds. Options identified by the IMF to address these gaps included giving regulators additional directions powers and the ability to levy the industry for any non-financial claims scheme (FCS)-related resolution expenditure by the authorities.

The previous Government consulted on measures to address most of the gaps, with little industry concern except on a small number of specific proposals. The Inquiry supports this continuing process.18 Although many of the gaps identified for change were relatively minor in isolation, their cumulative closing would enhance APRA’s crisis management powers and more closely align Australia with international standards and best practice.

Advertisement

Pre-planning and pre-positioning

Strong resolution powers are not enough. Regulators and Government need to be willing and prepared to use them. Many crisis management options are only credible with significant pre-planning. In a crisis, the more options available, the more likely a credible, low-cost option to prevent a disorderly collapse can be found that does not involve putting taxpayer funds at risk. Pre-planning can also increase the consistency of Government approaches to crises and, through public communication, can increase the predictability and transparency of Government responses.

Australia could do more to be in a position to use resolution powers effectively. This could involve pre-planning for failure, both generically and by developing plans for specific institutions, and testing these plans through crisis simulations.19 The regulators can oversee that institutions build credible sources of loss absorbency in resolution and, at the same time, Government can ensure its balance sheet remains strong.

Pre-planning is not without cost. It can be resource intensive for regulators. It can also impose a burden on the industry, which may have to devote resources to develop internal recovery plans, provide data to regulators and make business changes to address any identified barriers to resolution. However, compared to other options, additional pre-planning is likely to be relatively low cost.

Advertisement

Capital requirements

In December 2013, APRA identified the four major banks as domestic systemically important institutions and increased their capital requirements, from 2016, by 1 percentage point.20 Increased capital requirements for systemically important banks are in line with international practice. Most jurisdictions adopting the Basel framework have introduced, or will introduce, similar measures. Australia’s requirement is at the low end of the international spectrum, which ranges from 1 percentage point to around 6 percentage points for the largest banks in Switzerland (Chart 5.1). In their submission, the regional banks suggested that a higher capital add-on for systemically important banks could be warranted.

Chart 5.1: Capital ratio add-ons for systemically important banks(a), (b)

This chart shows the capital ratio add-on for systemically important banks in a number of jurisdictions. Australia has the equal lowest capital add-on at 1 percentage point, while Switzerland has the highest with a range of 1 to 6 percentage points.

(a) Includes capital add-ons for both Global Systemically Important Banks (G-SIBs) and Domestic Systemically Important Banks (D-SIBs). For the European Union this is for ‘other systemically important institutions’.

(b) Where countries have a range of possible capital add-ons, ‘minimum’ and ‘maximum’ show the floor and ceiling of that range. Where countries have the same capital add-on for all systemically important banks, this is showed as its ‘minimum’.

Sources: APRA, BCBS, China Banking Regulatory Commission, De Nederlandsche Bank, Swiss Financial Market Supervisory Authority (FINMA), Hong Kong Monetary Authority, Japan Financial Services Agency, Monetary Authority of Singapore, Office of the Superintendent of Financial Institutions Canada, Reserve Bank of India, Sveriges Riksbank.

Increasing capital requirements reduces the likelihood of institutional failure. It gives a greater capital buffer to systemically important banks, whose collapse would cause significant damage to financial markets and the economy. Higher capital also helps ameliorate the effects generated by perceptions of an implicit guarantee.

Some stakeholders have argued Australian banks already have adequate levels of capital. ADIs have increased capital levels since the GFC and have arguably reduced the risk in their asset portfolios.21 In addition, equity funding is typically thought to be more expensive than debt funding, although a greater use of equity funding reduces bank failure risk and therefore may lower investors’ required return on equity and the cost of borrowing.22

The Inquiry notes that the FSB’s framework for global systemically important institutions includes insurers as well as banks. The IAIS is currently finalising the methodology for identifying global systemically important insurers, although this is unlikely to include any Australian institutions. It is not yet clear what arrangements, if any, will be made for domestic systemically important insurers.

Advertisement

The Financial Claims Scheme

The FCS was introduced as part of the Government’s response to the GFC. It provides a government guarantee of retail deposits held at Authorised Deposit-taking Institutions (ADIs) up to the value of $250,000 per account holder per ADI. The scheme fully covers around 99 per cent of eligible depositors and over half of deposits by value.23 The deposit guarantee aims to give depositors confidence in the safety of their money during a crisis, preventing panic and bank runs that may exacerbate the crisis. A similar scheme is in place for general insurance policy holders should an insurer fail.

Like all government guarantees, the FCS could create moral hazard for depositors by removing credit risk.24 The depositor can place funds with any ADI without regard to the riskiness of the institution. In the context of reducing the adverse effects of too-big-to-fail, the FCS assists smaller institutions; absent such a guarantee, depositors could be expected to move their funds to the larger institutions if they perceive these to be the safest.

Some submissions criticise the form of the FCS, noting that pre-positioning for implementation is complicated and has been expensive. In particular, the ‘single customer view’ required per individual, per ADI has practical difficulties. In addition, complexity may make it challenging to distribute funds to affected depositors in a timely manner. This may undermine the scheme’s effectiveness in a crisis; it is important to meet depositors’ expectations of fast and accurate payouts, both to ensure system stability and to avoid losses for individuals. It should be noted that the increased confidence from the presence of the FCS may assist in preventing a crisis from occurring.

The FCS threshold of $250,000 is high by comparison to most deposit insurance schemes internationally,25 and some submissions argue it is too high.26 This threshold should be high enough to cover the average depositor’s funds to avoid people withdrawing funds during a crisis. However, the higher the threshold, the greater the allocative efficiency distortion it can cause, giving greater incentive for individuals to invest in deposits compared to other assets.27

Currently, the FCS is post-funded. This means that, if it were activated, the Government would initially pay out claims and then recover those funds from the assets of the failing institution. If that was not sufficient, the remainder would be recovered through a levy on the rest of the banking sector, which could be delayed until the crisis was over to avoid exacerbating the situation.

An option is to charge ADIs an ex ante fee, or pre-funding, for the FCS; the IMF recommended that Australia “Re-evaluate the merits of ex ante funding for the FCS”.28 Ultimately this fee would likely be passed on to depositors, which would satisfy a ‘user pays’ principle for the deposit insurance provided. In addition, depending on how it was structured, an ex ante fee could collect a dedicated pool of funds that could be rapidly accessed to meet FCS claim needs. Broadening the allowable use of these funds to include measures that reduce the magnitude of FCS claims — such as assisting in resolution — may reduce the overall burden of the scheme.

However, an ex ante model would impose a cost on the financial sector. In particular, industry would need to pay a fee that would likely be passed on, at least in part, to depositors in the form of lower deposit interest rates or higher fees. This would be the case even if there was no need to activate the FCS. By contrast, the current ex post funding only imposes a cost on industry if the guarantee is needed.

Advertisement
Advertisement

Ring-fencing

Following the GFC, several major jurisdictions introduced or proposed significant structural reforms to their banking sectors through ring-fencing. The goal of ring-fencing is to ‘carve out’ specific financial activities to protect them from other activities that are less critical to economic activity — and are likely to be riskier. This might involve separating commercial banking from investment banking, or insulating domestic operations from risks in offshore activity. There is no one way to implement ring-fencing, with approaches differing by country (Box 5.1).

Such measures help address the costs of too-big-to-fail institutions in three ways:

  1. Although all parts of a bank may collectively be too-big-to-fail, each individual part may not be. This is not a matter of size — the complexity of an institution also affects whether it is possible to resolve in an orderly fashion. A simpler internal bank structure, where core activities are already separate, would make resolution easier.
  2. Although there may be a political and economic imperative to provide government support to core services, such as access to retail deposits, this is less likely to be the case for investment banking activities. Ring-fencing can allow Government greater ability to limit support to only the core aspects of the business, reducing the associated perceived implicit guarantee.
  3. By separating different types of activities, an institution’s approach to risk appetite, including remuneration structures which affect risk taking, may become more appropriately aligned to those activities.

Ring-fencing in Australia

Currently, the share of Australian banks’ balance sheets used for investment banking activities, and the extent of proprietary trading, is lower than in many other jurisdictions.29 Among other things, this could reflect a lack of retail competition that makes ‘plain vanilla’ banking profitable without the need to take on riskier business, or relative returns to investment banking in Australia being low.

Ring-fencing would come at a cost. Of the measures discussed in this section, it is likely to be the most burdensome. It involves costs to institutions of restructuring, ongoing efficiency costs through reduced diversification benefits, and may introduce barriers to foreign entrants and limit Australian banks’ ability to expand internationally.

The United Kingdom Treasury estimates ring-fencing United Kingdom institutions will involve a transition cost of around £3 billion, with ongoing costs of around £420 million to £1.9 billion per annum (0.04 to 0.16 per cent of United Kingdom GDP). Efficiency can also be a casualty, as firms are forced into a corporate structure other than what they would choose if not constrained. However, the estimated net present benefit of the United Kingdom reforms is £114 billion through reduced probability and severity of future financial crises.30

Unlike in the United Kingdom, the mixture of retail and investment banking in Australia is more limited. This means the immediate costs and benefits of such a policy may also be more limited. That said, introducing ring-fencing now could assist in avoiding future issues if banks were to move more into riskier activities. For example, banks may move into riskier investment bank activities searching for higher returns, as increased competition or other factors make safer lending less profitable.

Although implementing ring-fencing now would be costly, it would be less costly than if it were introduced at a time when banks were engaged significantly in both ring-fenced and other activities.

It is worth noting that Australia had a type of de facto ring-fencing before the 1990s, when the major banks each had separate trading bank and savings bank arms, but Government policy during deregulation removed that distinction.

The Inquiry seeks views on whether ring-fencing could be of net benefit to Australia and, if so, what model of ring-fencing would best suit our conditions. In particular: what types of activity should be ‘inside’ the fence and what should not; and how ‘high’ should the fence be — that is, how strongly should activities be separated?

Advertisement
Advertisement
About the author
David Llewellyn-Smith is Chief Strategist at the MB Fund and MB Super. David is the founding publisher and editor of MacroBusiness and was the founding publisher and global economy editor of The Diplomat, the Asia Pacific’s leading geo-politics and economics portal. He is also a former gold trader and economic commentator at The Sydney Morning Herald, The Age, the ABC and Business Spectator. He is the co-author of The Great Crash of 2008 with Ross Garnaut and was the editor of the second Garnaut Climate Change Review.